ANTS AND PLANTS 193 



Chinese, the most expert of agriculturists, recognize this 

 fact. In a Chinese orange-grove it is quite usual to 

 see lengths of bamboo connecting some of the trees, 

 and these are placed to serve as bridges to conduct 

 ants from ant-infested trees to trees infested by de- 

 structive insect pests. If ants were as abundant in the 

 orange-groves of Florida as they are in the tropics, 

 we should perhaps hear less of the ravages of that 

 very destructive pest, the Orange-Scale. Nevertheless, 

 I believe that the benefits conferred on myrmecophilous 

 plants have been greatly over-rated, and I regard some 

 of the wonderful developments of plant structure as 

 protective devices against the too pressing attention 

 of ants, rather than as devices to attract ants. If one 

 plant can inherit the power to produce a definite 

 pathological structure, I do not see why another can- 

 not inherit the power to produce a structure adapted 

 to harbour with least damage to itself guests that 

 certainly will not fail to make their appearance. 



We may now pass to the consideration of the other 

 plants of my list. 



Macaranga caladifolia is a little shrubby plant about 

 3 feet high ; the stem is hollow, slightly swollen, 

 and here and there perforated by little holes. The 

 leaves are peltate, with veins running out to the 

 margin in a radiate manner, and those at the base of 

 the leaf terminate on the margin each in a large 

 glandule which secretes a sweet fluid. Ants are 

 found in the hollow stems. 1 



1 The story of the truly myrmecophilous species of Macaranga 



in the Malay Peninsula is described by H. N. Ridley in the Annals 



of Botany, XXIV. (1910), pp. 470-83. There are two series, in both 



of which the stem, at first solid, becomes hollow by dilatation 



14 



